基于小波变换及融合技术的图像边缘提取

图像边缘检测是图像分割、图像识别的基础,传统边缘检测是基于图像整体的边缘检测.小波变换使基于图像分解的边缘检测成为可能,利用小波变换将数字图像分解为高频和低频分量,对高频和低频分量分别进行边缘检测.常规融合方法是将高频边缘和低频边缘进行简单叠加,由于高频、低频边缘是通过不同方法提取的,二者之间的相似度、吻合度存在差异,简单叠加不能够有效融合高频、低频边缘图像特征,本文算法采用局部区域方差准则把高频和低频边缘在小波域进行融合.实验表明,该算法能够有效融合高频、低频边缘图像特征,具有较好的边缘检测和去噪功能.

Edge Extraction Based on Wavelet Transform and Fusion

Edge detection is the basis of image segmentation and image recognition.Traditional detection methods are carried out on whole image.With the new algorithm for image edge detection which is based on image decomposition,the digital image is decomposed into a high-and a low-frequency images using wavelet transform,then edge detections are carried out to these two images respectively.The normal method is just to add the high-frequency edge and the low-frequency edge.Because there are differences between the two detection images,the fused image can not combine their characteristics effectively.In this paper,the algorithm with the standard of region variance makes them fused in wavelet area,a new edge image with a better quality is obtained.The result of experiments indicates that the algorithm can fuse the characteristics of high-and low-frequency edge images effectively,and has good functions for image edge detection and image de-noising.